Abstract
Alzheimer’s Disease (AD) is a neurological disorder that affects cognitive functions, including memory, thinking, and behavior. Early detection of Alzheimer’s disease is critical for effective treatment and management of the condition. Deep Learning (DL) is a powerful tool that can be used for AD detection and diagnosis. DL algorithms can learn patterns and features in large datasets that can be used to classify and predict the presence of Alzheimer’s Disease. The most common approach is to use brain imaging techniques, such as computed tomography and brain MRI scans, to extract features that are characteristic of Alzheimer’s Disease. Transfer learning-based deep learning models can be effective in detecting Alzheimer’s disease from medical images. Transfer learning involves using pre-trained neural network models as a starting point and fine-tuning them to suit a specific task, such as Alzheimer’s disease detection. This paper focuses on classifying AD patients into various stages (early mental retardation, mild mental impairment, late mild mental impairment, and final Alzheimer’s stage) by utilizing transfer learning with ResNet50, VGG16, and DenseNet121 along with CNN networks on a large dataset. The work classifies Alzheimer’s patients into various stages using transfer learning with ResNet50, VGG16, and DenseNet121 along with CNN on a large dataset. The model is trained and tested on ADNI data using Keras API and divides the MRI images into: EMCI, MCI, LMCI, and AD. The performance of VGG16, DenseNet121, and ResNet50 outperformed other models significantly. The results demonstrate a significant improvement in accuracy compared to previous approaches, with a final accuracy of 96.6%.